Butterflies. II: Torsors for 2-group stacks (Q990753): Difference between revisions

This is the second of a series of articles considering the higher-dimensional analogues of torsors for \(n\)-group stacks. This article focuses on 2-group stacks and the corresponding torsors. A 2-group is a 2-groupoid with a single object. In other words, it is a 2-category in which both 1-cells and 2-cells are invertible. It corresponds to a group object in the category of small categories, to a crossed module, to a suitably truncated simplicial group, and so on, so there are a lot of interacting structures to call on when working with them. The innovation of this series of articles is to concentrate attention on the weak morphisms between 2-groups, or, more exactly, between the 2-group stacks that were studied in detail in the first paper of the series [\textit{E. Aldrovandi} and \textit{B. Noohi}, Adv. Math. 221, No. 3, 687--773 (2009; Zbl 1179.18007)]. (Such weak morphisms can be usefully represented by `butterfly' diagrams, whence the title of the series.) The paper is very readable, but is a sequel so some acquaintance with Part 1 is assumed. The scope of the paper includes discussion of torsors and non-abelian cohomology, gerbes bound by a crossed module (an idea due to \textit{R. Debremaeker} and \textit{K. U. Leuven} [Bull. Soc. Math. Belg., Sér. B 29, 57--72 (1977; Zbl 0421.18006)], extensions of gerbes along a butterfly and the links between this approach and the Schreier-Grothendieck-Breen theory of extensions.